using the source code directly with some editing in my project

[jaraheel]

Hi,
I tried installing your library using Julia package manager, but that gives some compatibility issues with the DataFrames, Flux, and a few other libraries' versions I have installed. To circumvent this issue, I downloaded your source files and added them to my project with some editing. Is this ok? I would credit you in the published code and documentation for my project.

[darrencl]

Hi @jaraheel,

Thank you for using this package! I haven't worked on this for a while. May I ask if your project is a private project or you'll make it public?

I would rather fixing t his issue. I have created a PR (#6) to fix this and will merge and do minor release once passed the CI. The project itself runs fine, but I need to fix the documenter CI. I will let you know once it's done so you can update your version of FeatureSelectors.

In the meantime, if you wish, you can use this version by:

(@v1.7) pkg> add FeatureSelectors#bump-deps

Thanks,
Darren

[jaraheel]

Hi Darren,
My project would eventually be public. It is for academic, non-commercial purposes.

[jaraheel]

just attempted to add the library with #bump-deps, and it still conflicts with DataFrames:

ERROR: Unsatisfiable requirements detected for package DataFrames [a93c6f00]:
DataFrames [a93c6f00] log:
├─possible versions are: 0.11.7-1.3.2 or uninstalled
├─restricted to versions 1.3.0-1 by FeatureSelectors [e0e7d8aa], leaving only versions 1.3.0-1.3.2
│ └─FeatureSelectors [e0e7d8aa] log:
│ ├─possible versions are: 0.1.0 or uninstalled
│ └─FeatureSelectors [e0e7d8aa] is fixed to version 0.1.0
└─restricted by compatibility requirements with LowRankModels [15d4e49f] to versions: [0.11.7-0.19.4, 0.21.0-0.21.8] — no versions left
└─LowRankModels [15d4e49f] log:
├─possible versions are: 1.0.0-1.1.1 or uninstalled
└─restricted to versions * by an explicit requirement, leaving only versions 1.0.0-1.1.1

[darrencl]

Hi @jaraheel,

Thank you for trying that! I think I just fixed it by adding more flexibility with caret specifier in the compat. Could you try re-adding FeatureSelectors#bump-deps again?

Below is the example of my environment in Julia 1.7. I've able to install this alongside with DataFrame 1.3.2.

(@v1.7) pkg> status
      Status `~/.julia/environments/v1.7/Project.toml`
  [a93c6f00] DataFrames v1.3.2
  [e0e7d8aa] FeatureSelectors v0.1.1 `https://github.com/darrencl/FeatureSelectors.jl.git#bump-deps`
  [587475ba] Flux v0.12.9
  [09f84164] HypothesisTests v0.10.6
  [ce6b1742] RDatasets v0.7.7
  [2913bbd2] StatsBase v0.33.16

Hope this fixes the issue!

[darrencl]

@jaraheel Seems like the CI passed. I will squash merge tomorrow then release minor version tomorrow.

The current version of bump-deps branch should work for you, let me know how that goes.

[jaraheel]

it still gives me the same error
pkg> add FeatureSelectors#bump-deps
Updating git-repo https://github.com/darrencl/FeatureSelectors.jl.git
Resolving package versions...
ERROR: Unsatisfiable requirements detected for package DataFrames [a93c6f00]:
DataFrames [a93c6f00] log:
├─possible versions are: 0.11.7-1.3.2 or uninstalled
├─restricted to versions 1.3.0-1 by FeatureSelectors [e0e7d8aa], leaving only versions 1.3.0-1.3.2
│ └─FeatureSelectors [e0e7d8aa] log:
│ ├─possible versions are: 0.1.1 or uninstalled
│ └─FeatureSelectors [e0e7d8aa] is fixed to version 0.1.1
└─restricted by compatibility requirements with LowRankModels [15d4e49f] to versions: [0.11.7-0.19.4, 0.21.0-0.21.8] — no versions left
└─LowRankModels [15d4e49f] log:
├─possible versions are: 1.0.0-1.1.1 or uninstalled
└─restricted to versions * by an explicit requirement, leaving only versions 1.0.0-1.1.1

[darrencl]

Hi @jaraheel

I think your issue is not with FeatureSelectors.jl anymore. DataFrames.jl has already released v1.* and it seems that LowRankModels.jl locks it to the older versions.

I just checked and found this issue (madeleineudell/LowRankModels.jl#120). Someone has submitted a PR (madeleineudell/LowRankModels.jl#121) to fix this. Let's hope this will get merged soon.

All in all, you should use latest released version of DataFrames.jl. So, what you need to do is:

• Remove LowRankModels.jl, i.e. from repl ]remove LowRankModels
• Update your DataFrames.jl, i.e. ]up DataFrames. Ensure you have latest version installed by checking package status ]status DataFrames
• Add LowRankModels.jl from the PR commit hash ]add LowRankModels#e15afec

Below is an example in my environment. It fails to install LowRankModels.jl released version because it locks DataFrames to older version (while I have newer version installed and my package require newer version, since there were some breaking changes since 0.21 — version required by released LowRankModels.jl).

(@v1.7) pkg> add LowRankModels
   Resolving package versions...
ERROR: Unsatisfiable requirements detected for package LowRankModels [15d4e49f]:
 LowRankModels [15d4e49f] log:
 ├─possible versions are: 1.0.0-1.1.1 or uninstalled
 ├─restricted to versions * by an explicit requirement, leaving only versions 1.0.0-1.1.1
 ├─restricted by compatibility requirements with StatsBase [2913bbd2] to versions: 1.1.0-1.1.1 or uninstalled, leaving only versions: 1.1.0-1.1.1
 │ └─StatsBase [2913bbd2] log:
 │   ├─possible versions are: 0.24.0-0.33.16 or uninstalled
 │   └─restricted to versions 0.33 by FeatureSelectors [e0e7d8aa], leaving only versions 0.33.0-0.33.16
 │     └─FeatureSelectors [e0e7d8aa] log:
 │       ├─possible versions are: 0.1.1 or uninstalled
 │       └─FeatureSelectors [e0e7d8aa] is fixed to version 0.1.1
 └─restricted by compatibility requirements with DataFrames [a93c6f00] to versions: uninstalled — no versions left
   └─DataFrames [a93c6f00] log:
     ├─possible versions are: 0.11.7-1.3.2 or uninstalled
     └─restricted to versions 1.3.0-1 by FeatureSelectors [e0e7d8aa], leaving only versions 1.3.0-1.3.2
       └─FeatureSelectors [e0e7d8aa] log: see above

Using that commit from PR solves the issue since it's now able to use newer DataFrames.jl version. It also seems to pass their CI, so it should be good to go and safe for you to use.

(@v1.7) pkg> add LowRankModels#e15afec
     Cloning git-repo `https://github.com/madeleineudell/LowRankModels.jl.git`
    Updating git-repo `https://github.com/madeleineudell/LowRankModels.jl.git`
   Resolving package versions...
   Installed RandomizedLinAlg ─── v0.1.0
   Installed Arpack_jll ───────── v3.8.0+0
   Installed NMF ──────────────── v0.5.3
   Installed NLSolversBase ────── v7.8.2
   Installed FiniteDiff ───────── v2.11.0
   Installed NonNegLeastSquares ─ v0.4.0
   Installed Parameters ───────── v0.12.3
   Installed Arpack ───────────── v0.5.0
   Installed Optim ────────────── v1.6.2
  Downloaded artifact: Arpack
    Updating `~/.julia/environments/v1.7/Project.toml`
  [15d4e49f] + LowRankModels v1.1.1 `https://github.com/madeleineudell/LowRankModels.jl.git#e15afec`
    Updating `~/.julia/environments/v1.7/Manifest.toml`
  [7d9fca2a] + Arpack v0.5.0
  [6a86dc24] + FiniteDiff v2.11.0
  [d3d80556] + LineSearches v7.1.1
  [15d4e49f] + LowRankModels v1.1.1 `https://github.com/madeleineudell/LowRankModels.jl.git#e15afec`
  [d41bc354] + NLSolversBase v7.8.2
  [6ef6ca0d] + NMF v0.5.3
  [b7351bd1] + NonNegLeastSquares v0.4.0
  [429524aa] + Optim v1.6.2
  [d96e819e] + Parameters v0.12.3
  [85a6dd25] + PositiveFactorizations v0.2.4
  [0448d7d9] + RandomizedLinAlg v0.1.0
  [6e75b9c4] + ScikitLearnBase v0.5.0
  [3a884ed6] + UnPack v1.0.2
  [68821587] + Arpack_jll v3.8.0+0
Precompiling project...
  14 dependencies successfully precompiled in 12 seconds (118 already precompiled)

Hope this helps!

[jaraheel]

it worked! Thank you very much. Time to test :)